Aalborg Universitet

Frequency participation by using virtual inertia in wind turbines including energystorage

Xiao, Zhao xia; Huang, Yu; Guerrero, Josep M.; Fang, Hongwei

Published in:Proceedings of 43rd Annual Conference of the IEEE Industrial Electronics Society, IECON 2017

DOI (link to publication from Publisher):10.1109/IECON.2017.8216419

Publication date:2017

Document VersionEarly version, also known as pre-print

Link to publication from Aalborg University

Citation for published version (APA):Xiao, Z. X., Huang, Y., Guerrero, J. M., & Fang, H. (2017). Frequency participation by using virtual inertia in windturbines including energy storage. In Proceedings of 43rd Annual Conference of the IEEE Industrial ElectronicsSociety, IECON 2017 (pp. 2492-2497). IEEE Press. https://doi.org/10.1109/IECON.2017.8216419

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

? Users may download and print one copy of any publication from the public portal for the purpose of private study or research. ? You may not further distribute the material or use it for any profit-making activity or commercial gain ? You may freely distribute the URL identifying the publication in the public portal ?

Take down policyIf you believe that this document breaches copyright please contact us at vbn@aub.aau.dk providing details, and we will remove access tothe work immediately and investigate your claim.

Frequency Participation by Using Virtual Inertia

in Wind Turbines Including Energy Storage

Xiao Zhaoxia1, Huang Yu1, Josep M. Guerrero2

1.Tianjin Key Laboratory of Advanced Technology of Electrical Engineering and Energy

Tianjin Polytechnic University, P. R. China, Email: xiaozhaoxia@tjpu.edu.cn

2. Dept. of Energy Technology, Aalborg University, Denmark, joz@et.aau.dk www.microgrids.et.aau.dk

Abstract—With the increase of wind generation penetration,

power fluctuations and weak inertia may attempt to the

power system frequency stability. In this paper, in order to

solve this problem, a hierarchical control strategy is

proposed for permanent magnet synchronous generator

(PMSG) based wind turbine (WT) and battery unit (BU). A

central controller forecasts wind speed and determines

system operation states to be sent to the local controllers.

These local controllers include MPPT, virtual inertia, and

pitch control for the WT; and power control loops for the

BU. The proposed approach achieve at the same time both

output power smoothing and frequency participation.

Simulation results confirm the effectiveness of the proposed

control strategy in different scenarios.

Keywords—Wind power, virtual inertia control, frequency

regulation, microgrids, automatic generation control (AGC)

I. INTRODUCTION

The penetration of the wind power installed is

increasing worldwide, however impacting seriously the

grid performances [1]-[4]. Wind power fluctuations

may result in frequency instabilities, especially in

weak-grid situations. Recently, more permanent

magnet synchronous generator (PMSG) based wind

turbines (WT) are becoming more popular in wind

farms duo to the higher performances required in new

grid code. In contrast to the conventional synchronous

generators, PMSG, because of the use of back-to-back

converters, disturbances in the wind generator little

affect the main grid. However, the WT rotor speed is

decoupled from the grid frequency, which may result in

a lack of inertia, i.e. frequency changes will not produce

power variations, as it is desired in a power system [5]-

[6].

Further, placing energy storage systems in wind

power systems can effectively smooth wind power

fluctuations, and can improve the output controllability

of the wind power generation, which has great

significance to the frequency regulation [7]-

[8].Relevant research has been done on wind

generation frequency regulation [9]-[10]. In order to

make WTs to participate in frequency regulation, we

can use power reserve and rotor inertia controllers. A

method to do that consist on reserving some amount of

power, so that the power generation is out from the

optimal value. Another way to achieve it is to reduce

the output active power by using pitch control, thus

when the frequency drops, reserved power may be used

to support the system frequency. However, the

abovementioned methods are at the expenses of

reducing wind power efficiency.

Alternatively, rotor inertia control can be also used

providing a short time power support for DFIG-based

WTs [9]. In [10], frequency-rotor speed coordinated

control strategy is adopted to reduce sudden WT torque

changes, and to adjust rotor speed when the frequency

changes. However, in [11] was pointed out that rotor

speed recovery may cause the secondary frequency

drop down.

To solve this problem, in [12] by introducing energy

storage may help wind power systems in frequency

regulation. In this work, a flywheel is used in

combination with a DFIG for these purposes. In [13],

BU is connected to the ac bus of the stationary

synchronous compensator, while in [14] a super

capacitor is be connected to the dc bus of PMSG wind

power generator. Although the abovementioned

methods may be used to make wind power participating

in frequency regulation, they present complex

implementation and control.

In order to solve wind power and frequency

fluctuations, this paper uses a method to calculate the

expected interchanged power flow of the wind power

system tie-line, according to the predicted wind speed

inside a determined time period. The designed WT-BU

structure connects the BU to the ac part through an

inverter. A power-frequency loop is added PMSG side

converter, so that the system inertia is effectively

increased. When the system frequency exceeds

specified range, the wind power generation system

ensures the MPPT performances, while the WT-BU

active power output is rapidly changed according to the

automatic generation control (AGC) instructions. A

hierarchical control structure is presented to implement

those functionalities. Simulation results and analysis

are given showing, the performances of the proposed

approach.

Generation

Consuption

Loca

l con

trollers

MPPT control

of WT

Pitch angle

control

Rotor speed

recovery control

Grid side

control of WT

Constant

power

control of

battery

Virtual inertia

control

Main grid

PMSG

A

B

C

PMSG

...

Reference setting of local

controller

Selection of operation

states

Central controllers

Da

ta a

cqu

isition

Battery

Wind

turbine 1

Wind

turbine n

Pmpp

Pmpp

isabc

Iq*

Udc Disp

atc

h c

enter

Igabc

PLL

Ugabc

UwabcIwabc

IBabc

Iwabc

Selection local

controllers

Pf

Fig. 1. Structure of WT-BU system.

II. HYBRID STRUCTURE OF WT-BU SYSTEM

Fig. 1 shows the hybrid WT-BU system, including the power

stage and the control architecture. In this case the BU is

connected to the AC side to facilitate the retrofitting. Note that

adding batteries in the DC side may be more effective, but in

practice it may not be suitable since it may require cooperation

between WT and BU producers companies, which may not be

easy in many cases. Since in our case, the hardware control

between the WT and the energy storage system is completely

decoupled from each other, it is more feasible in practice.

Regarding the controllers, the easiest way is to have WT-

BU locally controlled and to add a central controller that can

be placed in a SCADA system that may control the hybrid

power system. The central controller includes wind

forecasting, power calculations, system operation mode

selections, battery charge/discharge management, and the

real-time grid frequency detection.

Through the coordinated control of WT-BU, when the

frequency of the system is nominal, the power supplies

smoothed power to the main grid. Furthermore, when the

frequency of the system is outside the normal range, the WT-

BU power system participates in the frequency regulation.

III. FREQUENCY PARTICIPATION

A. Kinetic Energy of the Wind Turbine

The kinetic energy stored in the rotor blade of a WT can

be expressed as:

E =1

2Jω2 (1)

where J is the inertia of the wind turbine and its generator and

𝜔 is the rotor speed. When the rotor speed is ranged from 𝜔0

to 𝜔1, the difference of the kinetic energy of the wind turbine

is：

∆E =1

2Jω0

2 −1

2Jω1

2 (2)

From (2), we can conclude that the wind power system has

a certain rotational kinetic energy. Conventional variable-speed

WT, in order to achieve maximum wind energy capture, uses

back-to-back converters in order to decouple rotor speed and

grid frequency. Therefore, a frequency-dependent power

command 𝑃𝑓 can be imposed on the active power reference

value of the rotor side converter controller.

Hence when the system frequency is nominal, 𝑃𝑓 will be

zero, otherwise 𝑃𝑓 will change accordingly. This way, the

kinetic energy stored in the wind power system can be

converted into electrical energy in order to support the grid

frequency.

B. Battery participation in power grid frequency regulation

The proposed WT-BU hybrid power system is able to follow

AGC power scheduling PAGC to restore the grid frequency, see

Fig. 2. For instance, when the grid frequency is out from the

nominal range, the WT-BU hybrid system will react according

to the AGC scheduling instructions in order to quickly change

its active power output thus achieving frequency restoration.

The system reacts according to the following strategy to control

the frequency. If the grid frequency is high, BU is charging,

thus reducing the output power of the tie-line. Otherwise, if the

grid frequency is low, BU is discharging, thus increasing the

output power of the tie-line.

PI-+

f

f * PAGC

Fig. 2. Block diagram of the automatic generation control (AGC).

IV. CENTRAL CONTROLLER

The central controller takes care of the BU management

and the different operation modes of the WT-BU power system.

A. Output power reference of BU inverter

The wind energy generated 𝑊𝑊 during the scheduling

period T can be expressed as follows:

WW = ∫ PWdtT

0 (3)

being PW the forecast wind power generation.

Thus, during this scheduling period T, the average wind

power, PL, takes the form:

PL =WG

T (4)

If the grid frequency is nominal, the expected total active

power output during the scheduling period T is the output power

average over this period. At this time, the reference output

power of the BU inverter can be expressed as:

Pref = PL − PG (5)

where PG is the actual output power of WT.

If the main grid requires all WTs to participate in the

frequency regulation, the expected total active power

output during the scheduling period T is the output power

average plus the active power of frequency regulation

PAGC over this period. The reference output power of the

BU inverter can be expressed as:

Pref = PL + PAGC − PG (6)

Notice that the output power of BU, PB, should track Pref.

B. Operation modes and transition conditions

The central controller takes care of managing the different

operation modes. Notice that V is the actual wind speed, while

Vn is the rated nominal wind speed; f is the actual grid frequency,

fmin and fmax are the minimum and maximum grid frequency.

Battery maximum power charge and discharge are noted as PBC

and PBD.

Mode 1: If V ≤ Vn, the frequency will be nominal, WT

operates in the MPPT mode, and the tie-line active

power of is regulated to a constant value. The reference

value of the active power of the energy storage inverter

is：𝑃𝑟𝑒𝑓 = 𝑃L − 𝑃𝐺.

Mode 2: If V>Vn, the frequency of the power grid is

nominal, the pitch control of wind generation system

starts working, to limit the output power of wind power

the active power output of wind-energy storage system

tie-line stays a constant value. The reference value of

the active power of the energy storage inverter is：

𝑃𝑟𝑒𝑓 = 𝑃L − 𝑃𝐺.

Mode 3: When 𝑓 > max and ∆P + 𝑃𝐴𝐺𝐶 ≥ 𝑃𝐵𝐶 (∆P =

𝑃𝐺 − 𝑃L), the energy storage starts participating in the

frequency regulation, 𝑃𝑟𝑒𝑓 = 𝑃L − 𝑃𝐺 + 𝑃𝐴𝐺𝐶.

Mode 4: When 𝑓 < min and ∆P + 𝑃𝐴𝐺𝐶 < 𝑃𝐵𝐷 (∆P =

𝑃𝐺 − 𝑃L )，and at this time rotor speed of wind

turbine ω ≤ 0.8𝜔∗, 𝜔∗ is rated speed of wind turbine,

the energy storage start to adjust the frequency, 𝑃𝑟𝑒𝑓 =

𝑃L − 𝑃𝐺 + 𝑃𝐴𝐺𝐶.

Mode 5: When 𝑓 < min and ∆P + 𝑃𝐴𝐺𝐶 < 𝑃𝐵𝑚𝑎𝑥, and at

the same time rotor speed of wind turbine ω > 0.8𝜔∗,

virtual inertia starts operating. When rotor speed

dropped to 0.7𝜔∗, the frequency is not restored to the

nominal value, and at the same time BU starts

participating in the frequency regulation 𝑃𝑟𝑒𝑓 = 𝑃L −

𝑃𝐺 + 𝑃𝐴𝐺𝐶.

It should be noted that when the frequency of the power

grid is too high or too low, the PAGC is greater than the BU

charge/discharge maximum output power, and the hybrid WT-

BU system is disconnected from the power grid, running in

islanded mode. Tables I present the operation modes and

transition conditions of the WT-BU system.

TABLE I. OPERATION MODES AND TRANSITION CONDITIONS

Mode State transition condition

1 V≤Vn

fmin≤f≤fmax

2 V> Vn && PG>PL

fmin≤f≤fmax

3 f> fmax && ∆P+PAGC≥PBC

(∆P=PG-PL)

4 f< fmin, ≤0.8*

∆P+PAGC≤PBD

5 f< fmin, >0.8*

∆P+PAGC≤PBD (∆P=PG-PL)

V. LOCAL CONTROLLER

In this Section, the local controllers of both WT and BU

are presented.

A. Wind generation controller

The PMSG based WT use a back-to-back PWM converter,

which avoids the use of gear box, thus improving system

efficiency and reliability.

1) PMSG side control

The controller includes MPPT algorthm, pitch control, and

inner current loops. In is well known that the wind energy

captured by wind turbines can be expressed as:

𝑃𝑤 =1

2𝜌𝜋𝑅2𝐶𝑝𝑣3 (7)

where ρ is air density,R is the radius of the wind turbine, and

𝐶𝑝 is the wind energy utilization factor, being a function of the

pitch angle β and the tip speed ratio λ，λ = ωR/v, while ω is

the speed of wind turbine. When the pitch angle is constant, the

WT runs at the optimum tip speed ratio, so that it can get the

maximum wind energy utilization factor 𝐶𝑝𝑚𝑎𝑥 . Therefore, at

a certain wind speed, the WT maximum power is only related

to the rotor speed:

𝑃𝑜𝑝𝑡 = 𝑘𝜔3 (8)

PI PI

dq

dqαβ

PI

2

πθp

abc SVPWM

kω 3

ω

θ

Ps

Pmpp

Pg

isd*=0

isd isq

iabc

isq*

Δusd

Δusq

usd usq

uα uβ

udc

Pw

Fig. 3. Block diagram of generator side control

k=1

2𝜌𝜋𝑅5𝐶𝑝

𝜔3

𝜆3 (9)

The PMSG rotor-side control block diagram is shown in

Fig. 3. The generator side converter is designed to achieve

maximum power tracking for wind energy by using stator flux

linkage vector control. By using the synchronous reference

frame transformation in dq axis, the stator voltage equation can

be expressed as:

{𝑢𝑠𝑑 = −𝑅𝑆𝑖𝑠𝑑 − 𝐿𝑠

𝑑𝑖𝑠𝑑

𝑑𝑡− 𝜔𝑒𝐿𝑠𝑖𝑠𝑞

𝑢𝑠𝑞 = −𝑅𝑆𝑖𝑠𝑞 − 𝐿𝑠𝑑𝑖𝑠𝑞

𝑑𝑡− 𝜔𝑒𝐿𝑠𝑖𝑠𝑑 + 𝜔𝑒𝜑

(10)

The electromagnetic torque can be obtained as:

𝑇𝑒 =3

2𝑝[𝜑𝑖𝑠𝑞 + (𝐿𝑠𝑑 − 𝐿𝑠𝑞)𝑖𝑠𝑞𝑖𝑠𝑑] (11)

where Rs is the stator resistance, Lsd and Lsq is the stator

inductance, 𝑖𝑠𝑑 and 𝑖𝑠𝑞 is the stator current, usd and usq is the

stator voltage, 𝜔𝑒 is synchronous angular velocity, φ is

permanent magnet flux linkage and 𝑝 is the number of pole

pairs. Considering that the current 𝑑-axis component is always

fixed to 0, then the electromagnetic torque can be expressed as

𝑇𝑒 =3

2𝑝𝜑𝑖𝑠𝑞 (12)

2) Grid Side Control

The aim of the grid side controller is to stabilize DC bus

voltage and to achieve P/Q decoupling control. Grid side

controller block diagram is shown in Fig. 4.

-++

-

+-

+-

+

++

abcdq

PI-

dqαβ

SVPWMPI

PI PI

ωLsq

Udc id*

iq*

iq

id

θ

Qw*

Qw

ωLsd

Udc* Ud*

Uq*

ΔUd

ΔUq

Ud

Uq

Uα Uβ

Fig. 4. Wind power grid side controller

The grid-side converter dq model can be expressed as:

{u𝑔𝑑 = −𝑅𝑔𝑖𝑔𝑑 − 𝐿𝑔

𝑑𝑖𝑔𝑑

𝑑𝑡+ 𝜔𝑔𝐿𝑔𝑖𝑔𝑞 + 𝑒𝑔𝑑

u𝑔𝑞 = −𝑅𝑔𝑖𝑔𝑑 − 𝐿𝑞𝑑𝑖𝑔𝑞

𝑑𝑡−𝜔𝑔𝐿𝑔𝑖𝑔𝑑

(13)

where egd andegq is the grid voltage, igd and igq is the grid

voltage, Rgand Lg is the resistance and inductance of grid-side

converter, and ωg is grid angular frequency.

The grid side controller uses grid P/Q calculation as:

{𝑃𝑔 = 𝑒𝑔𝑑𝑖𝑔𝑑 + 𝑒𝑔𝑞𝑖𝑔𝑞 = 𝑒𝑔𝑑𝑖𝑔𝑑

𝑄𝑔 = 𝑒𝑔𝑑𝑖𝑔𝑞 − 𝑒𝑔𝑞𝑖𝑔𝑑 = 𝑒𝑔𝑑𝑖𝑔𝑞 (14)

From (13), igd and igq references can be achieved by

independent controlling Pg and Qg.

3) Pitch control

When wind speed exceeds the rated wind speed, active

power output of the WT needs to be limited by changing the

wind turbine pitch angle β, hence reducing the wind energy

capture coefficient. The output power Pw of the current

generator is compared with the reference power Pw*. The error

is processed by the PI controller to produce the required β* (see

Fig. 5). Thus the pitch angle is changed accordingly, so that the

WT output power is kept at reference Pw*.

Pw*

PIβ*

+-

Pw

1/(TS+1)

Fig. 5. Pitch control.

4) Virtual inertia and rotor speed control

Fig. 6 shows the block diagram of the virtual inertia and

the rotor speed control. The virtual inertia controller detects

when the grid frequency drop down, making the rotor speed

decreasing, thus the rotating kinetic energy is released to

prevent further decreasing the grid frequency.

PI

+-

P

-+

1

2

0

Virtual inertia control

Rotor speed control

MPPT

+

fref

f

ref

P Pmpp

DP

Pf

Fig. 6. Rotor inertia control and speed recovery control

When starting the virtual inertia loop, the controller will

connect DP to switch 1 (see Fig. 6), and the power reference

will change according to the frequency deviation that will go

through a P controller that will added over the MPPT output

(PMPP). In order to prevent WT losing too much kinetic energy,

when the rotor speed dropped down to 0.7 times of the rated

value, the virtual inertia loop is disconnected, and DP will be

connected to switch 0. If the grid frequency is nominal, the

speed recovery control will start, connecting DP to switch 2.

Then, the rotor speed control will generate the power reference,

which will be obtained by using the rotor speed deviation

through a PI controller that will be subtracted from PMPPT.

B. BU Inverter Power Control

The BU inverter power controller structure is shown in

Fig. 7. The grid voltage uabc and inverter grid current iabc are

used to calculate P/Q, controlling those to follow Pref/Qref, and

then control Id/Iq through internal current loops.

SVPWMabc

dq

abc

ab

ab

dq

PLL

-ωL

ωL

PI

PI

+

+

+-

+-+

+

ab

dq

P=VgdId+VgqIq

Q=-VgdIq+VgqId

-+ -+

iabc vg_abc

Ɵ

Ɵ Vgd Vgq

Iq* Id*

Vgd

Vgq Iq*Iq Id

Iq

Id

Vgd

Vgq

Vga

Vgb

Qref

Pref

PI PI

P

Q

Id*

+

+

Fig. 7. Energy storage inverter power control

VI. MICROGRID CASE STUDY

The model of a hybrid WT-BU system connected through

a tie-line to a microgrid system was developed by using

Matlab/Simulink. The selection and switching of the operating

modes of the WT-BU system embedded in the central controller

is realized based on the stateflow toolbox. The main parameters

of the system are shown in Table II. The rated capacity of the

microgrid is 10MW, and the unit regulation power of the wind

storage system is 1MW/Hz. The nominal grid frequency is

fn=50 Hz, with +/-1% of error, so that the maximum and

minimum frequencies where fmax=50.5 Hz and fmin=49.5 Hz.

The microgrid control includes a secondary control based on

AGC to restore the frequency.

Wind speed curve, as shown in Fig. 8, cut the wind speed

at 6m/s, and the rated wind speed is 12m/s. According to the

above mentioned, according to the wind speed curve, the

calculation of the tie-line power is PL =940kW.

TABLE II SIMULATION PARAMETERS

Wind turbine parameters Value

Wind turbine radius /m 27.5

Rated wind speed m/s 12

Air density kg/m3 1.225

Rated power /MW 1.1

Best tip speed ratio 6.325

Wind energy maximum utilization factor 0.4385

Pitch control proportional coefficient 0.01

Permanent magnet wind turbine parameters Value

Rated power /kW 1.1

Flux linkage 4.74

Inertia /Joules 411258

Pole pairs 42

Stator resistance /Ω 0.006

d-axis inductance /mH 1.704

d-axis inductance /mH 1.216

Grid side converter parameters Value

Line inductance /mH 0.2

DC side capacitance /μF 60000

DC side voltage /V 1200

Inner loop regulator gain kp1, ki1 1, 10

Outer loop regulator gain kp2, ki2 2, 120

Rotor side converter parameters Value

Rotor side d-axis current inner loop kp3, ki3 25, 1

Rotor side q-axis current inner loop kp4, ki4 40, 9

Active power outer loop kp5, ki5 0.0032, 0.059

Rotor inertial control loop kp6, ki6 250000, 0

Speed recovery loop kp7, ki7 350000, 200000

Energy storage system parameters Value

Rated voltage of battery /V 1

Battery capacity /Ah 800

Battery maximum charge current 0.2C

Battery maximum discharge current 0.4C

Maximum discharge power PBD

Maximum charge power PBC

600 kW

-160 kW

PQ controls d-axis current inner loop kp8, ki8 2, 20

PQ controls q-axis current inner loop kp9, ki9 2, 20

PQ control active power outer loop kp10, ki10 0.001, 0.5

PQ control reactive power outer loop kp11, ki11 0.001, 0.5

Fig. 8. Wind speed curve

The simulation results of the wind storage system in

different operating modes are shown in Fig 9. F and f are

frequency of the wind-energy storage system before and after

the participation in the frequency regulation in Fig. 9 (a). Fig.

9 (b) is the wind generation output power; Figure 9 (c) is the

generator rotor speed; Figure 9 (d) is DC voltage of wind power.

Figure 9 (e) shows battery charging and discharging current.

Figure 9 (f) is battery output power; Figure 9 (g) shows the

active power output of WT-storage system.

(a) Frequency participation of the WT-storage system

(b) Output power of wind turbine

(c)Rotor speed

(d) DC voltage of wind power

(e) Battery charging and discharging current

0 2 4 6 8 10 12 14 16 180

5

10

15

t/s

m/s

2 4 6 8 10 12 14 16 18

49.5

50

50.5

t/s

Hz

f

F

2 4 6 8 10 12 14 16 18

-1

0

1

x 106

t/s

W

Pref

Pg

2 4 6 8 10 12 14 16 181

1.5

2

2.5

t/s

rad

<Rotor speed wm (rad/ s)>

2 4 6 8 10 12 14 16 180

500

1000

1500

t/s

V

2 4 6 8 10 12 14 16 18

-500

0

500

1000

t/s

A

(f) Battery output power

(g) Active power output of WT-storage system

Fig. 9. Simulation results.

VII. CONCLUSION

This paper proposed a grid-connected wind generation

system retrofitted with a battery energy storage unit. The

proposal includes inertia emulation by linking the rotor speed

with the frequency, taking into account speed constraints and

coordinating the actions between WT and BU.

The paper introduces a control architecture in order to achieve

a good integration of the WT-BU system with the power

system, thus achieving the following salient features:

· Output power smoothing

· AGC power schedule following

· Frequency regulation and participation

Simulation results shown the hybrid WT-BU power

system with the proposed control architecture operates

properly in the context of a small weak grid.

ACKNOWLEDGEMENT

This work was supported by the Tianjin Science and

Technology Support Program Key Project and National Natural

Science Foundation of China (15JCZDJC32100,

17JCZDJC31300 and 51577124).

REFERENCES

[1] Miao Fufeng, Tang Xisheng, Qi Zhiping, “Capacity optimization of energy storage participating to wind plant primary frequency regulation,” Advanced Technology of Electrical Engineering and Energy, 2016， (04):23-29.

[2] Lou Suhua, Yang Tianmeng, Wu Yanwu. Coordinated optimal operation of hybrid energy storage in power system accommodated high penetration of wind power. Automation of Electric Power Systems, 2016,(07):30-35.

[3] J. W. Choi, S. Y. Heo and M. K. Kim, "Hybrid operation strategy of wind energy storage system for power grid frequency regulation," in IET Generation, Transmission & Distribution, vol. 10, no. 3, pp. 736-749.

[4] Miao Fufeng, Tang Xisheng, Qi Zhiping, “Analysis of frequency characteristics of power system based on wind farm-energy storage combined frequency regulation,” High Voltage Engineering, 2015，(07)： 2209-2216.

[5] Wang Haijiao, Jiang Quanyuan, “An overview of control and configuration of energy storage system used for wind power fluctuation mitigation,” Automation of Electric Power Systems, 2014,(19):126-135.

[6] Tang Xisheng, Miao Fufeng, Tang Xisheng, “Survey on frequency control of wind power,” Proceedings of The Chinese Society for Electrical Engineering, 2014,(25):4304-4314.

[7] Wang Haijiao, “Analysis of frequency characteristics of power system based on wind farm-energy storage combined frequency regulation,” Advanced Technology of Electrical Engineering and Energy, 2012(14): 152-155.

[8] Miao Fufeng, Tang Xisheng, Qi Zhiping. Analysis of frequency characteristics of power system based on wind farm-energy storage combined frequency regulation.Advanced Technology of Electrical Engineering and Energy, 2012(14): 152-155.

[9] Jiang Ping, Xiong chuanhua. A control scheme design for smoothing wind power fluctuation with hybrid energy storage system. Automation of Electric Power Systems, 2013,(01):122-127.

[10] R. Xiang, X. Wang and J. Tan, "Operation control of flywheel energy storage system with wind farm," Proceedings of the 30th Chinese Control Conference, Yantai, 2011, pp. 6208-6212.

[11] N.Y.Abed, S. Teleke and J. J.Castaneda, "Planning and operation of dynamic energy storage for improved integration of wind energy," 2011 IEEE Power and Energy Society General Meeting, San Diego, CA, 2011, pp. 1-7.

[12] S. M. Muyeen, H. M. Hasanien and J. Tamura, "Reduction of frequency fluctuation for wind farm connected power systems by an adaptive artificial neural network controlled energy capacitor system," in IET Renewable Power Generation, vol. 6, no. 4, pp. 226-235, July 2012.

2 4 6 8 10 12 14 16 18

-5

0

5

10x 10

5

t/s

W

2 4 6 8 10 12 14 16 18-1

0

1

x 106

t/s

W